Bacteroides fragilis is the leading cause of anaerobic bacteremia and intraabdominal abscesses. The capsular polysaccharide complex of the prototype strain, 9343, confers the abscessogenic properties of the organism. The capsular polysaccharide complex of 9343 is comprised of at least eight distinct capsular polysaccharides (PSA1 - PSH1). For strain 9343, PSA1 is not only crucial for abscess formation, but is the only polysaccharide necessary for abscess formation by this organism. The region of the B. fragilis chromosome containing the PSA1 biosynthesis locus is heterogeneous, therefore, the PSA molecules synthesized by different B. fragilis strains are structurally distinct. It is not known whether the PSA molecules of other virulent strains also confer the abscessogenic potential of that organism. The zwitterionic nature of PSA1 of 9343, having both a positive and negative charge per repeating unit, is essential for its abscessogenic potential. Two genes of the PSA1 biosynthesis locus, wcfR and wcfS, are conserved in the PSA biosynthesis loci of all strains analyzed. Homology-based data suggest that the products of these genes are involved in the formation of the positively charged monosaccharide of PSA1, AATGaI. Therefore, it is likely that all B. fragilis strains synthesize a PSA molecule with this same positively charged monosaccharide (AATGal). Based on the importance of AATGal to the virulence of the PSA1 of 9343, our overall hypothesis is that the PSA molecule confers the abscessogenic potential of each virulent strain. This application is divided into three aims that will address this hypothesis, first at the species level by mutating the PSA loci of various strains that produce structurally distinct PSA molecules and testing these mutants for their ability to induce abscesses. Next, the genes and their products that are predicted to be involved in the synthesis of AATGal will be analyzed biochemically and genetically. We now have the tools and an adequate scientific foundation to determine why the species B. fragilis as a whole has abscessogenic capabilities rather than only understanding this phenomenon for the prototype strain. The data gained from these aims may allow us to realize our goal of elucidating the genetic basis of abscess formation by this species.